A method including in response to voice information, acquiring a response object to which corresponding voice information is registered and which occupies an interactive medium; and informing the response object to respond to the voice information. The technical solution enables an operating system to automatically determine a response object responding to voice information triggered by a user from numerous response objects in view of voice information registered at each response object and the occupation of interactive media by each response object. The technical solution removes the restriction that a user must speak a wakening phrase to activate a required response object each time before speaking voice information. Thus, a user only needs to speak short voice information that shows the intention for use thereof during a process of using a certain response object, thereby improving the convenience of a user performing voice interaction.